Separated by a Common Language

Native New Yorker, and if I said it at all (way to pretentious a saying) I would say "if needs be" and heard it more with the "s". The other version was much less common in my hearing.

"Needs must" sounds incredibly awkward to my AmE ears, though I have heard the apparently equivalent "must needs" (mostly in Shoghi Effendi’s translations of the Bahá’í Writings--it is used four times in this 40-page section of the Book of Certitude alone.) Shoghi Effendi, though, was a Oxford-educated Persian, who purposely employed a slightly archaic English style in order to convey the loftiness of the original Persian and Arabic.

If need is a noun, the form of the subjunctive present is be in both singular and plural:

I be, you be, need be, we be, you be, needs be.

Indeed, but my point was that since most BrE speakers do not have a productive subjunctive in their dialects, they are likely to be (AmE) weirded out by it, and so maybe it sounds better to them in the plural. Just a guess, and it doesn't work terrifically well in this context, since the subjunctive form is the same as the base form (i.e. untensed form) of the verb--but there are lots of other subjunctive forms, including the past tense that applies to this phrase, that share the same form as the indicative plural.

Like dan, I (AmE) am not entirely comfortable with "needs must." Although I recognize the phrase, it rings BrE or possibly archaic to my ears.

"If needs be" is almost entirely unfamiliar -- although I might have heard it and just put it down to error. Just more anecdotal evidence.

I've tried to find an authoritative source for this, but I can't, so here's the hearsay: a Mormon linguist and net.friend told me that Utahn English was moderately influenced by BrE due to the large number of Britons who emigrated with the Latter Day Saints during the original colonization. He spoke primarily of phonology, but it's not surprising if certain idioms were also transferred.

Thanks John C, but it's unclear whether if needs be goes as far back as the original Mormon relocation to Utah in the mid-19th century. The OED has no examples, and the earliest I've found so far is in the 1980s. I can try a(n) historical corpus when I'm at my office...

Just checked the British National Corpus (most sources from 1991), and that has 13 instances of "if needs be" and 119 of "if need be"--so the s-less form is nearly 10 times more common in BrE in that corpus. But searching the Guardian website, there are 99 instances of "if needs be" yet only 288 of "if need be"--so the s-less form is less than three times more common on that site. We could take this as evidence (though we should check a more diverse set of sources first!) that the s-ful form is spreading (rather quickly) these days. The Guardian website seems to mostly cover articles since 1999, but there are some older articles included (probably because they're historically interesting). There are examples for "needs" from 1999, but none earlier, but we can't take much from that since so few articles from before 1999 are archived.

Semi-related, but is it my imagination or are 'do' forms less common in the US than in the UK? I'm particularly thinking of expressions like "I could do" or "They might do" where the 'do' acts as a pro-verb. American friends seem to find them odd.

And then there's ""I could do a pie".

Bingley, I have no idea why you think that topic is related to this one (even tangentially!)...but you happen to have raised the exact topic that I've been struggling with sentence trees for! So, I beg your patience (let's not discuss it yet!) and will try to get that post up as soon as I can.

Too many exclamation (BrE) marksM
/(AmE) points in this comment!!!

I always figured 'if needs be' was the correct form because of assumed words left out -- that the longer version was 'if (it) needs (to) be,' with the extra words being understood by the hearer to theoretically be there.

I'm surprised you didn't use the entire phrase which begins with "needs must" that I'm familiar with: Needs must when the devil drives! Surely this didn't die out in the Regency?

Interesting hypothesis, Eloise.

There are other versions of the 'devil drives' saying, with the 'needs must' word ordering not seeming to be fixed until sometime after Shakespeare--starting with "c1500 (?a1475) Assembly of Gods 21 He must nedys go that the deuell dryues" (OED, 2003)--but I didn't mention it because I didn't think it was terribly relevant to my hypothesis!

LG, can you give the evidence that shows that "need be" predates "needs be"? Not that I think you're lying -- I'd just like to see it.

BrE DOES have the older form. needs < OE nedes, adv. 'of necessity'. (Not sure if a macron would show if I included one, but imagine one on the first E, i.e. the first vowel was long.)

My MA thesis was on grammaticalization of 'must'-like modals, and if I hadn't come across 'needs must' back then, I would have been puzzled too.

Tangentially related to this thread: I've been trying, with limited success, to popularise a Day-Todayism. "More proof, if proof be need be, that..."

Do your bit!

If the s-less form is indeed the original one, 'folk grammar' could be responsible for the presumably incorrect inference of 'if (it) needs (to) be'. Part of the reason for this may well be that in a trend towards fewer and simpler rules modal verbs (which 'need' IS in the subjunctive 'need be' and IS NOT in 'needs to be') are increasingly replaced by non-grammaticalized verb forms, hence the spreading of the s-form.

Lynne, could you possibly parse "needs must" for me? I understand the meaning and I know it's a very old phrase in the language, but I've wondered for years about the (archaic, I assume?) grammar rules behind it. BTW, I'm quite sure I've only heard it/seen it in BrE.

James, my only evidence is that there are no citations of 'if needs be' in the OED--either under the phrase 'if need be' nor under the separate entry for 'needs'. I'm trying to look it up again on-line, but the OED seems to be feeling ill at the moment. I have had a look at my hard-copy 1971 OED, but the entries there are not at all as developed for the phrase 'if need be' as the 2003 revisions in the on-line version were...so I can't tell at the moment how far back their examples of 'if need be' go.

So, I should go back and edit the entry slightly to make it clear that I'm acting on absence of evidence rather than positive evidence.

Dreas, nedes, as found in the needs must phrase is indeed an older form, but what is the evidence that that was the form used in 'if need be' as a phrase? The OED (still on the old 1971 reprint) has an example of the noun needs-be from 1881 ("Peter could discover no needs-be in the humiliation and death of Christ"), but examples of the noun need-be from 1728 ("He afterward saw a remarkable Providence in it, and a Need-be for it"). Now, the noun need(s)-be could very well be the source of the modern phrase if need(s) be, but there's still no evidence here that this came directly from the older form nedes. What may have happened is that nedes survived in some dialect longer, and that's why we find late needs-bes, but if the phrase was coined in a need- rather than needs-saying dialect, then the needs form would still be newer, even though based on an old-fashioned word.

So, we just don't know. And so I will adjust the last paragraph of the entry a bit.

Julie, it may be a bit futile to try to parse such a set-phrase; people who use it these days are most likely treating it as a sort of lexicali{s/z}ed unit and not parsing its innards. Earlier examples are often must needs. If I were to try to parse it, it would be with needs acting in a sort of adverbial way (meaning 'of necessity'--OED) and a null subject. So, one can translate it into parsable contemporary English as something like 'It must necessarily be'.

I agree with Eloise, I am an AmE native, and have always said "If Needs Be" In fact, I am just the opposite of most of teh AmE speakers in that WITHOUT the "S" sounds odd to me.

Eloise's theory is exactly how I thought about the phrase.
But of course the other side of that argument is that the sentence could easily be "(you) need (to) be" as opposed to "(it) needs (to) be"
Perhaps because "Needs Be" more often than not talks about an intangible "it" (for example, the requirement to move a table if it starts to rain) than a you or I.

I'm a born and raised Californian and over here we say "if needs be".

A question for those who say 'if need_ be'. Do you feel 'need' is a verb, a noun, or an adverb?

Lynne, what do you think it is historically?

It's a noun, subject of the verb be.

I speculate: isn't 'needs be' simply an 'eggcorn'? A popular conflation of the proverbial 'needs must [when the Devil drives] ...' and '[If] need be'?

If this is the answer, there's no worry about how the subjunctive is conjugated!

Wasn't that my hypothesis 2?

Sorry, Lynne! By the time I read through all the comments, I must have forgotten! Didn't mean to plagiarize.

Just for the sake of completeness: the New York Times has 498 hits for the s-ful "if needs be", many (perhaps most...I didn't look that closely) of them from the 1800s and early 1900s. Three hits are from the past twelve months. This very quick, very unscientific search does tell us at least two things, though (I think): (1) the s-ful form can be found, relatively recently, in AmE journalistic prose; and (2) the form is not a recent BritE innovation. If anything, it's been lost in AmE. In any case, it sound quite odd to my AmE ears.

To my mind, in the phrase "if need be", be = exist, so the phrase means "if a/the need exists."
Sweet and simple, except I'm never sure about that period-inside-or-outside-the-quotes thing.
Enjoying the discussion!

Just for the record, I'm from New England, and I would use 'if need be' myself, but I have also heard 'if needs be,' and it doesn't sound completely wrong.

Second, I'm wondering how trustworthy newspapers are as a measure of common usage. The problem I see with newspapers is that they are overseen by editors who are interested in achieving a consistent written presentation. Individuals' word/grammar preferences may also be subjected to rules of usage that are created for the publication. Thus a few people could theoretically influence or change the way that many other people write (or at least are published).

Massachusetts-

I'm more comfortable with if need be than if needs be, but in my own speech I'm far more likely to say if necessary.

The way I read the OED, needs was originally an adverb in needs must. They comment:

Now freq. taken to be a plural noun and verb.

Well, needs must is part of my vocabulary and I personally don't take it to be PLURAL NOUN+ VERB. It's impossible to prove anyway, since must is not marked for singular or plural concord.

In any case, the fact that needs has a long history as an adverb makes If needs be a plausible

If [it] needs be [so] = 'If it necessarily is so'.

It's interesting that needs used to be common as an adverb after a modal verb. OK, a little archaise, but not so archaic as to prevent Kenneth Graham from writing for children in 'Wind in the Willows'

His spirits finally quite restored, he must needs go and caress his possessions, and..show off their points to his visitor.

Late to this thread, though I find the explanation proffered by Mr Crosbie for 'needs' as an adverb provocative. Could there be support for parsing this 'needs' as 'need' +genitive 's', in which case equivalent to 'of need' ('of necessity'), that is, then, a prepositional phrase used as an adverb here in 'needs must'?

Anonymous

I find the explanation proffered by Mr Crosbie for 'needs' as an adverb provocative

It's not me that's doing the provoking. The evidence collected by the OED is overwhelming.

• In Old English the adverb nedes could appear with any part of the clause.

• Since the 14th century, the word has been linked to the vern must or the earlier form mote, both before and after the verb. It's hard to see how the use must needs could involve a subject. (Look again at that quote from Wind in the Willows.)

• Needs must as a stand-alone phrase or preceding when the Devil drives etc has been around since the sixteenth century. The earliest quote in the OED after if is Robert Browning in 1871: She shall go, if needs must.

• Need was also used with will and would from the 14th century. Two quotes from famous writers:
Shakespeare He needes will be Absolute Millaine. (= 'absolute ruler of Milan')
Hobbes It would needs follow, not onely that all men were wicked..but also wicked by nature.

Your explanation of needs as a reduced prepositional phrase is implausible. Yes, the Old English noun was used with used with the propositions of, for and on. But that use petered out around 1600 in Middle English. Around that time the phrase of needs appeared, but the OED judge it to be 'now archaic and rare'.

The formation of adverbs with the suffix -(e)s was quite common in Old English Some of the adverbs (and prepositions) that have survived are discussed in the thread toward(s) and other ward(s). Another survivor is always.

In case anyone's wondering what happened to the post about the American subjunctive: It's in The Prodigal Tongue! From the book: "This return of the subjunctive is one of the most startling things to happen to English in centuries. Where did it come from? Why in America? Why then?" Buy it and find out!

(I'd say more, but comments on an old post are not the right place to start up a new discussion)

I grew up in New Jersey and always used "need be".
I recently wrote a clause "no condition need be applied", and someone wrote that it should be "needs", but to my ear, "need" sounds correct.